搜档网
当前位置:搜档网 › CANCER_RESEARCH

CANCER_RESEARCH

Protein Kinase C f Is Required for Epidermal Growth Factor–Induced Chemotaxis of Human Breast Cancer Cells

Ronghua Sun,1Ping Gao,3Lin Chen,1Dalong Ma,2Jiming Wang,3Joost J.Oppenheim,3

and Ning Zhang1,3

1Department of Chemical Biology and State Key Laboratory of Molecular Dynamic and Stable Structures,College of Chemistry and

2Laboratory of Medical Immunology,School of Basic Medical Science,Peking University,Beijing,China;and3Laboratory of Molecular Immunoregulation,Center for Cancer Research,Intramural Research Support Program,National Cancer Institute,Frederick,Maryland

Abstract

Chemotaxis plays an important role in cancer cell metastasis. In this study,we showed that epidermal growth factor(EGF) was a more potent chemoattractant than chemokine SDF-1a/ CXCL12for human breast cancer cell MDA-MB-231.Different inhibitors were used to evaluate the involvement of12protein kinase C(PKC)isotypes in the chemotactic signaling pathway. Chelerythrine chloride,an inhibitor of all PKC isotypes, blocked chemotaxis,whereas inhibitors of classic and novel PKC,such as Go¨6976,Go¨6850,or calphostin C,only impaired EGF-induced chemotaxis to a minor extent by e32%inhibi-tion.These data suggested that atypical PKC were involved. The ligand-induced actin polymerization and cell adhesion were also similarly dependent on atypical PKC.Immunofluo-rescent staining showed an EGF-induced,LY294002-sensitive translocation of PKC f from the cytosol to the plasma membrane,indicating that EGF was capable of activating PKC f,probably via phosphoinositide3kinases.A myristoy-lated PKC f pseudosubstrate blocked the chemotaxis with an IC50of20A mol/L.To expand our investigation,we further showed that in MCF-7and T47D,two additional human breast cancer cell lines,EGF-activated PKC f and the PKC f pseudo-substrate,inhibited chemotaxis.Taken together,our data suggest that PKC f is an essential component of the EGF-stimulated chemotactic signaling pathway in human breast cancer cells.(Cancer Res2005;65(4):1433-41)

Introduction

Chemotaxis,the directional movement of a cell in response to a chemical gradient,plays an important role in development, wound healing,angiogenesis,and immune responses(1–5). Recent studies indicate that chemotaxis contributes to the spread of cancer cells(6,7).Blocking of a chemokine receptor, CXCR4,impaired the breast cancer metastasis to regional lymph nodes and the lungs(6).Numerous studies have revealed that the molecular mechanism of chemokine receptor–orchestrated chemotaxis can involve G-protein-coupled receptors(8–11). Binding of a chemokine activates G i protein,resulting in a chain of downstream signaling events that include dissociation of heterotrimeric G proteins;activation of phosphoinositide3 (PI3)kinase g,Cdc42/Rac,and Arp2/3;and polymerization of actin at the leading edge of a cell(12–15).

Chemotaxis of breast cancer cells is also mediated by G-protein-independent receptors,such as epidermal growth factor(EGF) receptors,members of the receptor tyrosine kinase(RTK)family (16).In the presence of EGF,dimerization and autophosphoryla-tion of EGF receptors induce the exposure of cytosolic binding sites to several parallel downstream signaling molecules,such as phospholipase C g(PLC g),Grb2/Sos,and PI3kinases(17,18). However,the precise molecular mechanism of EGF-elicited chemotaxis is largely unknown.Recent studies suggest that either PI3kinase a or y is required(19,20).Blocking PLC g with U73122is reported to interfere with EGF-mediated breast cancer migration and tumor invasion(21–23).Cdc42is required for directional movement of NR6fibroblast and A431cells(24).Actin polymer-ization elicited by EGF also plays a critical role in cell migration (25).These data indicate that although EGF may share some of the molecular components,such as Cdc42and filamentous actin (F-actin),with chemokine-induced chemotaxis,it also utilizes a unique signal transduction pathway to direct cell migration. Both G-protein-coupled receptors and RTK activate members of the protein kinase C(PKC)family,consisting of12serine/threonine kinases.The PKC family can be divided into three subfamilies based on their activation mechanism:classic PKC,such as PKC a,h I,h II,g, which require both diacylglycerol and calcium for activation;novel PKC,such as y,q,u,D,the activation of which requires diacylglycerol and not calcium;and atypical PKC,including f and E,which do not need diacylglycerol or calcium for their activation(26,27).PKC plays pleiotropic roles in cell polarity,migration,and adhesion.For instance,phosphorylation of integrin a6h4by PKC a is associated with EGF-induced cell adhesion and migration(28).PKC a has also been shown to directly bind to h1integrin(29).However,a recent study shows that phorbol12-myristate13-acetate–induced PKC a activation results in a decrease in EGF-induced cell motility in MDA-MB-231cells(30).PKC q has also been shown to be colocalized with integrin h1and to contribute to cell motility(31).PKC y is required for EGF-induced phosphorylation of myosin light chain,a key step in forming contractile ring during migration(32).PKC f has been indicated to play an important role in determining cell polarity (33–37).Studies using isozyme-specific inhibitors suggest that PKC f is also involved in chemokine-triggered cell adhesion and actin assembly in polymorphonuclear cells,presumably mediated by G i protein(38,39).However,the role of PKC f in RTK-mediated chemotaxis has not been characterized.In the current study,we investigated the roles of PKC isotypes in EGF-induced chemotaxis and found that PKC f was the only essential PKC isotype. Materials and Methods

Cell Culture and Reagents.Human breast cancer cell line MDA-MB-231was obtained from American Type Culture Collection(Manassas,VA). T47D and MCF7were from Invitrogen(Carlsbad,CA).They were cultured

Requests for reprints:Ning Zhang,Department of Chemical Biology and State Key Laboratory of Molecular Dynamic and Stable Structures,College of Chemistry,Peking University,Beijing100871,P.R.China.Phone:86-10-62755154;Fax:86-10-62751708; E-mail:zhangnz@https://www.sodocs.net/doc/a89928815.html,.

I2005American Association for Cancer Research.

in RPMI1640with10%FCS(complete medium).Chemotaxis chambers and membranes were purchased from Neuroprobe(Gaithersburg,MD). Human EGF was from BD Biosciences(San Jose,CA).Fibronectin(0.1%) was from Sigma(St.Louis,MO).LY294002,U73122,AG1478,and AG17 were from Calbiochem(La Jolla,CA).Calphostin C,Go¨6850,Go¨6976,and chelerythrine chloride were from Alexis Biochemicals(San Diego,CA). Polyclonal antibodies against PKC f(sc-216),PKC a/h I/h II(c-20),and a monoclonal anti-human EGF receptor(EGFR)inhibitory antibody were from Santa Cruz Biotechnology,Inc.(Santa Cruz,CA).Rabbit anti-human EGFR polyclonal antibody was purchased from Cell Signaling Technology, Inc.(Beverly,MA).Oregon Green514phalloidin was from Molecular Probes,Inc.(Eugene,OR).Myristoylated and nonmyristoylated pseudo-peptides for PKC f were synthesized by Calbiochem based on the reported sequence from the pseudosubstrate region of human PKC f(SIYRR-GARRWRKL).Myristoylated scrambled peptide(RLYRKRIWRSAGR)was synthesized by GL Biochem(Shanghai)Ltd.(Shanghai,China).Peptides were solubilized immediately before use at a1mM concentration in PBS, pH7.2.

Chemotaxis Assay.Chemotaxis assays were done as described by the manufacturer(Neuroprobe,Cabin John,MD)and by Zhang et al.(40). Briefly,chemoattractants were loaded into the lower chemotaxis chamber. MDA-MB-231cells(0.5?106cells/mL),suspended in binding medium (RPMI1640,0.1%bovine serum albumin,and25mM HEPES),were then added into the upper chambers.The two chambers were separated by a 10A m filter that had been pretreated with0.001%fibronectin in RPMI1640 at4j C overnight.The chambers were incubated in5%CO2at37j C for 3hours.The filter was then washed,fixed,and stained.The number of migrated cells was counted in three high-power fields(400?).The chemotaxis index was calculated as the ratio of the number of cells responding to a chemoattractant gradient over the number of migrated cells in a medium control.For checkerboard assay,MDA-MB-231cells were mixed with different concentrations of EGF before adding to upper wells of the chamber.For inhibition by anti-EGFR inhibitory antibody,MDA-MB-231 cells were preincubated with indicated concentrations of anti-EGFR antibody or isotype-matched mouse IgG2a at room temperature for 15minutes.For the inhibitor assay,MDA-MB-231cells were pretreated with inhibitors at the indicated concentration for45to60minutes at37j C and then loaded into the upper chamber.Statistical analysis was carried out to determine the significance of chemotactic response using PRIZM3, ANOVA analysis.

Protein Kinase C Translocation Assay.Translocation of PKC in MDA-MB-231cells was analyzed using immunofluorescent staining techniques with confocal microscopic analysis(40).In brief,cells were cultured48 hours before the experiment then starved in binding medium for3hours followed by stimulation with10ng/mL EGF at37j C for5to10minutes before fixation with4%formaldehyde.For inhibitor control,cells were treated with50A mol/L of LY294002at37j C for1hour before EGF stimulation.Cells were then permeabilized with0.2%Triton X-100in buffer (10mM HEPES,20mM KH2PO4,5mM EGTA,2mM MgCl2,Dulbecco’s PBS, pH6.8)and stained with polyclonal antibodies to various PKC isotypes followed by probing with FITC-labeled goat anti-rabbit antibody.Cells were visualized using a Zeiss LSM410inverted fluorescent confocal microscope (Carl Zeiss,Inc.,Thornwood,NY).

Western Blotting Assay.Western blotting assays were carried out as described by Zhang et al.(40).Proteins(20A g per lane)were separated by8%SDS-PAGE,transferred onto polyvinylidene difluoride membranes(Pierce,Rockford,IL),probed with rabbit anti-EGFR or anti-PKC f(1:1000)antibodies,and visualized using enhanced chemilu-minescence reagents.In Akt/PKB and extracellular signal-regulated kinase(ERK)1/2activation assay,MDA-MB-231cells were activated by 10ng/mL EGF for5minutes in the presence or absence of inhibitors. Then,the cells were lysed for Western blotting analysis with the use of antiphosporylated Akt/PKB,Akt/PKB,phosphorylated ERK1/2,and ERK1/2antibodies.

Flow Cytometry Assay.Cell surface expression level of EGFR was monitored by using a flow cytometer[fluorescence-activated cell sorter (FACS)].MDA-MB-231,MCF-7,or T47D cells(5?105cells per sample)were washed twice with ice-cold FACS buffer(Dulbecco’s PBS,1%FCS,5 mM EDTA,and0.1%sodium azide,pH7.4).The cells were stained with anti-EGFR monoclonal antibody for30minutes on ice.After three washes with ice-cold FACS buffer,FITC-labeled secondary antibody was applied for another30minutes on ice.The cells were washed twice and analyzed immediately by flow cytometry(BD Biosciences).For experiment with synthesized PKC f pseudosubstrate peptides,MDA-MB-231cells were pretreated with50A mol/L myr-pseudosubstrate,50A mol/ L non–myr pseudosubstrate,or medium alone at37j C for45minutes. The cells were washed twice with ice-cold FACS buffer before immunostaining.

F-Actin Content Assay.F-actin was quantified by methanol extraction of Oregon Green514/phalloidin–stained cells as described previously(25).Briefly,MDA-MB-231cells were plated and cultured for 18hours in complete medium followed by further culturing in binding medium for2hours.Cells were then treated with the indicated inhibitors at specific concentrations or binding medium for another hour followed by50ng/mL EGF stimulation at37j C for the indicated period of time.Cells were fixed,permeabilized,and stained in the dark with Oregon Green514phalloidin diluted in F-buffer(10mM HEPES,20 mM KH2PO4,5mM EGTA,2mM MgCl2,Dulbecco’s PBS,pH 6.8)at room temperature for30minutes.After five washes,bound phalloidin was extracted with methanol at4j C for90minutes and subjected to fluorescence analysis at511nm excitation and529nm emission.At the same time,an aliquot of cells were analyzed by a bicinchoninic acid assay(Pierce)to determine total protein in the sample.Fluorescence signals were normalized against total protein.Results were expressed as relative F-actin content,where

Fàactin D t=Fàactin0

?efluorescence D t=mg per mLT=efluorescence0=mg per mLTAdhesion Assay.Glass coverslips were coated with10A g/mL fibronectin in serum-free RPMI1640for2hours at37j C and air-dried before use. Monolayer of MDA-MB-231cells were trypsinized for5minutes at room temperature and stopped with culture medium,pelleted,and resuspended to 2.7?105cells/mL in complete medium.The cell suspension was placed in a 37j C5%CO2incubator for20minutes before the adhesion assay.Cells were treated at37j C for45minutes with inhibitors at the indicated concentrations or with medium alone.EGF or medium(10ng/mL)was added to each of the cell suspensions containing inhibitors or medium control.EGF-treated or nontreated suspension(1.5mL)was then immediately added to the appropriate35mm dish containing the slide.Cells were allowed to attach for5minutes before gently washing and fixing the slides.The number of attached cells was counted by light microscopy(200?).

Results

Using gradients of EGF or SDF-1a/CXCL12,we compared the chemotactic effect of EGFR with CXCR4,a chemokine receptor. Consistent with previous reports,SDF-1a/CXCL12,a specific ligand for CXCR4,induced the chemotaxis of MDA-MB-231cells in a dose-dependent manner(Fig.1A;ref.6).EGF-induced chemotaxis also exhibited a typical bell-shaped dose-response curve and the chemotaxis indexes were as much as3-fold more than those induced by CXCL12.Checkerboard analysis indicated that EGF also induced chemokinesis,an indication of enhanced cell motility(Fig.1B).As shown in Fig.1C,the inhibitory antibody to EGFR blocked the chemotaxis in a dose-dependent manner,indicating that the chemotactic activity of EGF requires EGFR(23).Furthermore,AG1478(0.5A mol/L),a specific inhibitor of EGFR activation,also inhibited chemotaxis,suggesting that EGFR dimerization and autophosphorylation is required(23).

Cancer Research

To assess whether these two types of receptors share certain chemotactic signaling components,we examined the effects of pertussis toxin,LY294002,and U73122.Pretreatment with 150ng/mL pertussis toxin,a specific inhibitor of G i protein,irreversibly inhibited SDF-1a /CXCL12-induced chemotaxis (Fig.1D )but not EGF-induced chemotaxis.In the presence of 30A mol/L LY294002,a specific PI3kinase inhibitor,both SDF-1a /CXCL12and EGF-induced chemotaxis were inhibited,sug-gesting that each required PI3kinases (Fig.1E ;refs.19,20).U73122(2A mol/L),a specific inhibitor of PLC,also blocked the chemotactic effect induced by either SDF-1a/CXCL12or EGF (Fig.1F ;refs.21–23).Overall,our data showed that EGF was a more potent chemo-attractant for MDA-MB-231cells than CXCL12and suggested that EGFR-mediated chemotaxis required PI3kinases and was G i protein independent.

To assess the role of PKC family members in EGFR-mediated chemotaxis,we investigated the inhibitory effects of a spectrum of specific PKC inhibitors.Chelerythrine chloride,a specific inhibitor that inhibits all the PKC isotypes with an IC 50of 660nmol/L,blocked EGF-induced chemotaxis in a dose-dependent manner (Fig.2A ).The combination of chelerythrine and LY294002also inhibited chemotaxis (data not shown).However,in the presence of inhibitors of classic or novel PKC,EGF still elicited marked chemotaxis.Go ¨6976,a specific inhibitor for classic PKC isotypes with an IC 50value ranging from 2.3to 6nmol/L,exhibited <20%inhibition at a concentration of 50nmol/L (Fig.2A ).Calphostine C,a specific inhibitor for diacylglycerol binding site,inhibited by 32%without any dose-dependence (Fig.2A ).Go ¨6850,another specific inhibitor for classic and novel PKC,inhibited chemotaxis only by 18%at 100nmol/L (Fig.2A ).Thus,classic or novel PKC

isotypes

Figure 1.EGF-induced chemotaxis of MDA-MB-231cells.A ,comparison of chemotactic dose-response of MDA-MB-231cells to EGF,CXCL12(P <0.0027).

B ,chemokinesis of MDA-MB-231cells induced by EGF.MDA-MB-231cells were mixed with indicated concentrations of EGF or medium only before immediate application to upper chamber.

C ,dose-dependent inhibition of EGF-induced MDA-MB-231cell migration was achieved by anti-EGFR antibody (P <0.0041)and EGFR-specific inhibitor AG1478(P <0.0072).MDA-MB-231cells (0.5?106cells/mL)were incubated with 1,2,and 5A g/mL of anti-EGFR-specific antibody or isotype-matched mouse IgG2a at room temperature for 15minutes and their chemotactic response to EGF at 10ng/mL was determined.

D ,effect on chemotactic dose-response to EGF or CXCL12stimulation following treatment with pertussis toxin at 150ng/mL (P <0.0051,two-way ANOVA).

E ,effect on chemotactic

dose-response to EGF or CXCL12stimulation following treatment with PI3kinase inhibitor LY294002at 30A mol/L.F ,effect on chemotactic dose-response upon EGF or CXCL12stimulation following treatment with 2A mol/L U73122.Chemotaxis assay was done as described in Materials and Methods.Columns,mean of triplicate measurements;bars ,SD.Chemotaxis data were analyzed by one-way or two-way ANOVA.

PKC f in EGF-Induced Chemotaxis

arenot essential for EGF-elicited chemotaxis.To confirm these observations,we assessed the effect of all these PKC inhibitors at doses of1to100ng/mL EGF.Once again,only treatment with chelerythrine chloride completely blocked EGF-elicited chemotaxis (Fig.2B).Therefore,this data suggests that an atypical PKC might be required for EGF-elicited chemotaxis.

Actin polymerization,a chemoattractant-induced response, provides an established indicator for the presence of a chemotactic signal.To confirm the results of the chemotaxis assays,a ligand-stimulated actin polymerization assay in the presence of selective inhibitors was done.Both EGF and SDF-1a/CXCL12induced a transient polymerization of globular actin,consistent with a previous report(Fig.3A;ref.25).Furthermore,we detected similar EGF-induced actin polymerization in MCF-7and T47D,two additional human breast cancer cell lines(data not shown).Our data also suggests that the EGF-and SDF-1a/CXCL12–induced signal transduction pathways converged at the point of actin polymerization.In the presence of30A mol/L LY294002or2A mol/L U73122,actin polymerization elicited by10ng/mL EGF was significantly reduced(Fig.3B),consistent with our chemotaxis data in Fig.1D and E.Treatment with chelerythrine chloride also inhibited actin polymerization,confirming that PKC was required for EGF-induced chemotaxis(Fig.3B).Chelerythrine chloride treatment also caused a decrease in the basal level of F-actin(data not shown).In the presence of Go¨6850,EGF still induced actin polymerization,further suggesting that classic and novel PKC isotypes were not essential for chemotaxis(Fig.3B).Taken together, our data implicates an atypical PKC in EGF-induced actin polymerization.Furthermore,the EGF-induced chemotaxis signal-ing pathway may converge with the G-protein-mediated pathway at or upstream of actin polymerization.

Adhesion is closely associated with cell migration and is another function frequently regulated by PKC.As shown in Fig.3C,10ng/mL EGF induced a rapid increase in cell adhesion to fibronectin surfaces.Go¨6850and Go¨6976did not interfere with the basal or EGF-stimulated cell adhesion.LY294002,U73122,and chelerythrine chloride each blocked EGF-stimulated rapid adhesion,indicating that PI3kinases,PLC,and an atypical PKC were involved. Treatment with chelerythrine C also inhibited basal adhesion, suggesting that an atypical PKC may also play a role in EGF-independent cell adhesion.

Because both actin and chemotaxis assays suggested that an atypical PKC might be involved in EGF-induced chemotaxis, we investigated whether EGF was capable of activating PKC f, an atypical PKC that has been implicated in modulating cytoskeleton structure.Subcellular redistribution is a hallmark of PKC activation(26).Using immunohistochemical staining, we examined the distribution of PKC f on EGF stimulation.As shown in Fig.4A,PKC f staining was distributed in the cytosol region of a resting cell.In the presence of10ng/mL EGF,the fluorescent signals were redistributed to the plasma membrane region of f34%of the MDA-MB-231cells,suggesting that

EGF

Figure2.Effects of PKC inhibitors on EGF-induced

chemotaxis.A,effect on chemotaxis of calphostin C,an

inhibitor of diacylglycerol binding,0-1.0A mol/L;Go¨6976,

an inhibitor of classic PKC,0-50nM;Go¨6850,an

inhibitor of both classic and novel PKC,0-100nM;and

chelerythrine chloride,an inhibitor of all PKC,0-5.0A mol/L

(P<0.0012).Cells were stimulated with10ng/mL EGF

following inhibitor treatment.B,effect on chemotactic

dose-response of different inhibitors.Cells were pretreated

with chelerythrine chloride(10A mol/L),calphostin C

(500nmol/L),Go¨6850(100nmol/L),and Go¨6976(50

nmol/L)before stimulation with EGF at various

concentrations(0-100ng/mL;P<0.0002).Chemotaxis

assay was done as described in Materials and Methods.

Columns,mean of triplicate measurements;bars,SD.

Chemotaxis data were analyzed by two-way ANOVA. Cancer Research

activated PKC f .In the presence of LY294002,this EGF-induced translocation of PKC f was impaired,suggesting that PKC f exerts its function downstream of PI3kinases in the chemotaxis signaling pathway.PKC a has been suggested to play a role in EGF-mediated chemotaxis.However,we were not able to detect a clear redistribution of cytosolic PKC a to the cytoplasmic membrane upon EGF stimulation.We further examined whether EGF activated PKC f of other human breast cancer cell lines.In 5minutes,EGF at 10ng/mL stimulated a similar redistribution of cytosolic PKC f to cell membrane in MCF-7and T47D as in MDA-MB-231cells (Fig.4B ).The translocation,sensitive to LY294002interference,occurred in 25%of MCF-7and 32%of T47D cells (data not shown).It seemed that cytosolic PKC f close to the cell membranes was among the first to be depleted.Thus,immunohistochemical data showed that EGF was capable of activating PKC f ,an atypical PKC in MDA-MB-231cells,and the activation process was PI3kinase dependent.

Akt/PKB,a pivotal effector directly downstream of PI3kinases,had been implicated in G-protein-coupled receptor–mediated chemotaxis (41).EGF induced Akt/PKB activation,as shown by an increase in phosphorylated Akt/PKB (Fig.4C ).Upon adding Akt inhibitor III,phosphorylation of Akt decreased up to 70%,indicating the inhibitor was functional (42).In the presence of Akt inhibitor III,EGF-induced chemotaxis was significantly impaired in a dose-dependent manner,suggesting that Akt was involved in EGF-induced chemotaxis (Fig.4C ).We further examined the role of ERK1/2in chemotaxis.PD98059,a specific inhibitor of ERK,induced a marked decrease in ERK1/2phosphorylation but had no detectable effect on EGFR-mediated chemotaxis.Thus,our data suggested that EGF-induced Grb2-Sos-ERK1/2pathway did not play a role in chemotaxis,consistent with previous report (23).

Finally,we used a PKC f pseudosubstrate,a proven specific inhibitor,to confirm the involvement of this molecule in EGF-induced chemotaxis (38,39).Myristoylation of this pseudosub-strate was required for delivering this peptide through the cytoplasmic membrane.In the presence of myristoylated peptide,EGF-induced chemotaxis was inhibited in a dose-dependent manner,with an apparent IC 50of 20A mol/L (Fig.5A ).There was no detectable change in cell motility or proliferation rate

after

Figure 3.Actin polymerization and cell adhesion of

EGF-stimulated MDA-MB-231cells.A ,time course of relative F-actin content in MDA-MB-231cells following EGF and CXCL12stimulation.EGF,50ng/mL;CXCL12,100ng/mL.B ,time course of relative F-actin content in MDA-MB-231cells following EGF stimulation with and without different inhibitors.

EGF,50ng/mL;chelerythrine chloride,10A mol/L;Go

¨6850,100nM;LY294002,30A mol/L;U73122,2A mol/L.F-actin content was determined as described in Materials and

Methods.Points,mean of triplicate measurements;bars,SD.C ,effect of EGF stimulation with and without different

inhibitors on adhesion of MDA-MB-231cells.EGF,10ng/mL;chelerythrine chloride,10A mol/L;LY29002,30A mol/L;Go

¨6850,100nM;Go ¨6976,50nM;U73122,2A mol/L.Adhesion assay was done as described in Materials and Methods.Columns,mean of triplicate measurements;bars,SD.The adhesion data were analyzed by one-way ANOVA (P <0.007).

PKC f in EGF-Induced Chemotaxis

addition of the peptide(data not shown).In controls,the nonmyristoylated peptide at50A mol/L caused no significant inhibition.Furthermore,the scrambled pseudopeptide did not impair chemotaxis,confirming the specificity of the pseudosub-strate inhibitor in use.To extend our investigation,two more human breast cancer cell lines,MCF-7and T47D,were examined for their EGF-induced chemotaxis.As shown in Fig.5B,EGF elicited a typical bell-shaped chemotactic response of these two cell lines,which was totally inhibited by myristoylated pseudo-substrate at50A mol/L.We next examined the expression pattern of PKC f.Western blotting data clearly showed robust expression of PKC f protein in all three human breast cancer cell lines,consistent with their critical role in chemotaxis(Fig.5C). One potential pitfall of the inhibitor experiments is that PKC f inhibitors down-regulates cell surface EGFR,resulting in a decrease in chemotaxis index.Western blotting and FACS analysis showed a considerable expression of EGFR on the surface of all three human breast cancer cells(Fig.5D and E). Treatment with myristoylated pseudosubstrate did not induce any detectable loss of surface EGFR,suggesting that PKC f pseudosubstrate is not involved in regulating the expression level of surface EGFR(Fig.5E).Taken together,our results indicated that PKC f is required for EGF-induced chemotaxis of human breast cancer cells.

Discussion

Accumulating reports have shown that PKC f plays pleiotropic roles in multiple signal transduction pathways,such as mitogen-activated kinase cascade,NF-n B activation,ribosomal S6-protein kinase signaling,and cell polarity pathway(26,33).Our data suggest that PKC f exhibits the novel function of mediating RTK-elicited chemotaxis.Previous reports have suggested that PKC a and PKC q are involved in EGF-induced chemotaxis(28–31).They enhance cell migration by directly phosphorylating integrins.It was even suggested that a phorbol12-myristate13-acetate gradient maybe sufficient to account for EGF-mediated cell migration(29).Our

data Figure4.Involvement of PKC f in EGF-induced cellular responses of MDA-MB-231cells.A,cellular distribution of PKC f and PKC a before and after EGF stimulation for10minutes.EGF,10ng/mL;LY294002,50A mol/L.Translocation assay was done as described in Materials and Methods.B,cellular distribution of PKC f in MCF-7and T47D cells at time0,5,and10minutes after10ng/mL EGF stimulation.C,in the presence of Akt inhibitor III at0,5,10,and20A mol/L,the phosphorylation of Akt/PKB–and EGF-induced chemotaxis were decreased.The total input of Akt was consistent(P<0.002).D,in the presence of PD98059at0,12.5,25,and 50A mol/L,the phosphorylation of ERK1/2is decreased.There was no detectable change in chemotaxis response induced by10ng/mL EGF.Total input of

ERK1/2was the same.(The chemotaxis data were analyzed by one-way ANOVA.)

Cancer Research

PKC f in EGF-Induced Chemotaxis

Figure5.A,blockage of MDA-MB-231cells chemotaxis in response to EGF by myristoylated PKC f pseudosubstrate peptide(myr-peptide).MDA-MB-231cells (0.5?106cells/mL)were incubated with10,20,and50A mol/L of myristoylated PKC f pseudosubstrate peptide,myristoylated PKC f scrambled peptide, nonmyristoylated PKC f pseudosubstrate peptide,or PBS(solvent for synthesized peptide)at37j C for45minutes.Then,the cells were measured for their chemotactic response to10ng/mL EGF or medium only.The results were shown as chemotactic index(F SD).Columns,mean of triplicate measurements;

bars,SD.B,inhibitory effects of50A mol/L PKC f pseudosubstrate on MCF-7and T47D human breast cancer cells.Columns,mean of triplicate measurements; bars,SD.Chemotaxis data were analyzed by one-way ANOVA(P<0.021).C,expression level of PKC f subtypes in MDA-MB-231,MCF-7,and T47D cells. Three breast cancer cell lines were collected separately and lysed.Twenty micrograms of cell lysates for each sample were subjected to8%SDS-PAGE and probed with specific rabbit anti-PKC f polyclonal antibody.D,Western blotting analysis of EGFR expression on MDA-MB-231,T47D,and MCF-7cells.Fifty micrograms of protein was loaded on each lane of SDS-PAGE.E,cell surface expression level of EGFR on MDA-MB-231,MCF-7,and T47D cells.Cells(0.5?106) of each cell line were collected and washed with FACS buffer.The cells were then stained with anti-EGFR monoclonal antibody or isotype-matched mouse

IgG2a followed by staining with FITC-conjugated rabbit anti-mouse antibody.F,effect of pretreatment with of PKC f pseudosubstrates on surface expression of EGFR by MDA-MB-231cells.MDA-MB-231cells(0.5?106cells/mL)were incubated with50A mol/L of myristoylated-PKC f pseudopeptide,nonmyristoylated pseudopeptide,or PBS only at37j C for45minutes.Then,the cells were washed twice with cold FACS buffer followed by immunostaining with anti-EGFR-specific monoclonal antibody or isotype-matched mouse IgG2a.Data showed percentage of positive staining cells.

argue that among various subtypes of PKC activated by EGF,PKC f plays an essential role in chemotaxis.The inhibitors of classic and novel PKC,such as Go¨6976,Go¨6850,or calphostine C,only impaired EGF-elicited chemotaxis to a minor extent.This minor inhibitory effect is probably caused by the blockade of PKC a or q,resulting in an impairment in EGF-induced phosphorylation of integrin(29,31). To examine whether a gradient of PKC activation accounts for the driving force of chemotaxis,we measured the chemotaxis index of MDA-MB-231cells in the presence of a phorbol12-myristate13-acetate gradient.We were not able to detect a significant chemotaxis in response to phorbol12-myristate13-acetate at0.02 to2000nmol/L(data not shown).When all12PKC subtypes were blocked by chelerythrine chloride,a specific inhibitor that interferes with the catalytic motif of all PKC,EGF-elicited chemotaxis was totally impaired,indirectly suggesting that atypical PKC isotypes may be the effectors participating in chemotaxis.Actin polymer-ization and cell adhesion assays further confirmed the essential role of atypical PKC in EGFR-mediated chemotaxis.The fact that EGF induced the translocation of PKC f to the plasma membrane directly indicates that EGF is capable of activating PKC f in MDA-MB-231 cells.Finally,the myristoylated pseudosubstrate of PKC f,a proven specific inhibitor,blocked EGF-induced chemotaxis in a dose-dependent manner.Thus,of the12isozymes,PKC f is uniquely required for EGF-mediated chemotaxis.We were able to confirm these observations in MCF-7and T47D,two additional human breast cancer cell lines,indicative of a general role of PKC f in human breast cell chemotaxis.

The molecular mechanism that governs EGF-induced PKC f activation still needs to be further elucidated.Immunofluorescent staining data suggests that PI3kinases are required for PKC f activation.Furthermore,inhibiting Akt/PKB,a signaling component immediately downstream of PI3kinases,impairs chemotaxis up to 50%.The lack of complete inhibition may indicate that activation of a small percentage of Akt/PKB is sufficient for chemotaxis or that an unidentified signaling molecule,downstream of PI3kinases,shares partially redundant function with Akt/PKB.Because PKC f does not contain a pleckstrin homology domain and cannot be directly activated by PI3kinases,we speculate that its activation relies on a pleckstrin homology domain–containing protein,such as Akt/PKB. Further analysis of the relationship between Akt/PKB and PKC f is under way by using siRNA that targets Akt/PKB.In the light of recent reports and our data,we propose three possible models explaining the mechanism of PKC f activation by PI3kinases(34–39).One pathway involves activation of PI3kinases that further stimulates Cdc42.Consequently,PKC f/Par6complex is recruited to the plasma membrane(34,35).Second,it has been shown that,upon activation by RhoA,PKC f is translocated to the plasma membrane and helps to enhance the lateral motility of LFA-1in lymphocytes (39).Thus,it is also possible that EGF induced PKC f activation through Rho.The third possibility is that phosphorylated PKC f binds to membrane-associated PDK1,a kinase that is recruited by the products of PI3kinases(33).Studies on the translocation of PKC f-GFP using a PKC f-GFP/MDA-MB-231cell line would further clarify pathway involved in EGF-induced chemotaxis.These studies are currently in progress.

Our data suggest that RTK-and G-protein-coupled receptor–mediated chemotaxis signaling pathways converge at or upstream of PKC f.In chemokine-induced chemotaxis,G h g dissociated from G a i activates PI3kinase g and PLC h II.PLC h II indirectly activates classic and novel PKC.However,PLC h II and novel or classic PKC do not play a significant role in G i-protein-mediated chemotaxis.Extensive research using wortmannin,LY294002,and pi3k cà/àmice has shown that PI3kinase g is essential in the signaling pathway(14,15). Its activity plays an important role in recruiting the downstream signaling molecules to the leading edge of a migrating cell(43). Recently,PI3kinase g has been shown to mediate lymphocyte adhesion and migration by regulating Rho and PKC f(39).EGFR seems to use different signaling molecules for chemotaxis.Upon ligand binding,EGF receptors dimerize to activate a spectrum of downstream signaling molecules,including Src homology and collagen protein,Janus-activated kinase1,PI3kinases,and PLC g (17,18).The fact that pertussis toxin does not interfere with EGF-elicited chemotaxis clearly rules out the involvement of G i-protein. However,similar to G i-mediated chemotaxis,activation of PKC f has been shown to be required for EGF-elicited human breast cancer cell chemotaxis,and the activation of PKC f seems to require PI3kinase as well.We speculate that PKC f is activated by PI3kinase a or subtypes.Taken together,our data suggest that although G-protein-coupled receptors and EGFR orchestrate different upstream signaling components,their pathways in mediating chemotaxis seem to converge at or before PKC f.Furthermore,because many RTK share similar signal transduction pathways,we further speculate that PKC f may be required for chemotaxis by other RTK. Identification of PKC f as a convergence point of EGFR and chemokine receptor–mediated chemotaxis provides us with a potential new target for anticancer drugs.Blocking CXCR4with antibodies effectively down-regulates cancer cell metastasis(6). However,there are many other receptors capable of inducing chemotaxis,such as CXCR1/2and EGF.In fact,our data show that EGF induces a greater chemotactic response than SDF-1a by MDA-MB-231cells.Thus,targeting one particular receptor may be insufficient to prevent cancer cell migration.Because PKC f is required by both EGFR and chemokine receptor–mediated chemo-taxis,it might be an effective target for more potent antimetastasis therapeutic strategies.Because the PKC f-deficient mice are grossly normal,pharmacologic inhibition of this enzyme should not be lethal for the hosts(44).We postulate that blocking PKC f may completely impair the chemotactic activities of some cancer cells, resulting in a decrease in tumor invasion and spreading. Acknowledgments

Received4/1/2004;revised9/28/2004;accepted11/18/2004.

The costs of publication of this article were defrayed in part by the payment of page charges.This article must therefore be hereby marked advertisement in accordance with18U.S.C.Section1734solely to indicate this fact.

We thank Edward Cho for technical support and Dr.Xinsheng Zhao for inspiring discussion.

References

1.Chen MY,Insall RH,Devreotes PN.Signaling through chemoattractant receptors in Dictyostelium.Trends Genet1996;12:52–7.

2.Dhawan P,Richmond A.Role of CXCL1in tumorigen-

esis of melanoma.J Leukoc Biol2002;72:9–18.

3.Locati M,Murphy PM.Chemokines and chemokine

receptors:biology and clinical relevance in inflamma-

tion and AIDS.Annu Rev Med1999;50:425–40.

4.Payne AS,Cornelius LA.The role of chemokines in

melanoma tumor growth and metastasis.J Invest

Dermatol2002;118:915–22.

5.Moser B,Loetscher P.Lymphocyte traffic control by

chemokines.Nat Immunol2001;2:123–8.

Cancer Research

6.Muller A,Homey B,Soto H,et al.Involvement of chemokine receptors in breast cancer metastasis. Nature2001;410:50–6.

7.Salcedo R,Martins-Green M,Gertz B,Oppenheim JJ, Murphy https://www.sodocs.net/doc/a89928815.html,bined administration of antibodies to human interleukin8and epidermal growth factor receptor results in increased antimetastatic effects on human breast carcinoma xenografts.Clin Cancer Res 2002;8:2655–65.

https://www.sodocs.net/doc/a89928815.html,er FI,Parent CA.PI3-kinases and PTEN:how opposites chemoattract.Cell2002;109:541–4.

9.Kutscher B,Devreotes P,Iglesias PA.Local excitation, global inhibition mechanism for gradient sensing:an interactive applet.Sci STKE2004;2004:13.

10.Kunkel EJ,Campbell DJ,Butcher EC.Chemokines in lymphocyte trafficking and intestinal immunity.Micro-circulation2003;10:313–23.

11.Condeelis JS,Wyckoff JB,Bailly M,et https://www.sodocs.net/doc/a89928815.html,mellipodia in invasion.Semin Cancer Biol2001;11:119–28.

12.Ma L,Rohatgi R,Kirschner MW.The Arp2/3complex mediates actin polymerization induced by the small GTP-binding protein Cdc42.Proc Natl Acad Sci U S A 1998;95:15362–7.

13.Welch MD,Iwamatsu A,Mitchison TJ.Actin polymerization is induced by Arp2/3protein complex at the surface of Listeria monocytogenes.Nature1997;385: 265–9.

14.Hirsch E,Wymann MP,Patrucco E,et al.Analysis of the murine phosphoinositide3-kinase c gene.Gene2000; 256:69–81.

15.Sasaki T,Irie-Sasaki J,Jones RG,et al.Function of PI3K g in thymocyte development,T cell activation,and neutrophil migration.Science2000;287:1040–6.

16.Grotendorst GR,Soma Y,Takehara K,Charette M.EGF and TGF-a are potent chemoattractants for endothelial cells and EGF-like peptides are present at sites of tissue regeneration.J Cell Physiol1989;139:617–23.

17.Bogdan S,Klambt C.Epidermal growth factor receptor signaling.Curr Biol2001;11:R292–5.

18.Schlessinger J.Cell signaling by receptor tyrosine kinases.Cell2000;103:211–25.

19.Hill K,Welti S,Yu J,et al.Specific requirement for the p85-p110a phosphatidylinositol3-kinase during epider-mal growth factor-stimulated actin nucleation in breast cancer cells.J Biol Chem2000;275:3741–4.

20.Sawyer C,Sturge J,Bennett DC,et al.Regulation of breast cancer cell chemotaxis by the phosphoinositide 3-kinase p110y.Cancer Res2003;63:1667–75.21.Xie H,Pallero MA,Gupta K,et al.EGF receptor

regulation of cell motility:EGF induces disassembly

of focal adhesions independently of the motility-

associated PLC g signaling pathway.J Cell Sci1998;

111:615–24.

22.Chen P,Xie H,Sekar MC,Gupta K,Wells A.

Epidermal growth factor receptor-mediated cell mo-

tility:phospholipase C activity is required,but

mitogen-activated protein kinase activity is not

sufficient for induced cell movement.J Cell Biol

1994;127:847–57.

23.Price JT,Tiganis T,Agarwal A,Djakiew D,Thompson

EW.Epidermal growth factor promotes MDA-MB-231

breast cancer cell migration through a phosphatidyli-

nositol3V-kinase and phospholipase C-dependent

mechanism.Cancer Res1999;59:5475–8.

24.Chou J,Burke NA,Iwabu A,Watkins SC,Wells A.

Directional motility induced by epidermal growth

factor requires Cdc42.Exp Cell Res2003;287:47–56.

25.Chan AY,Raft S,Bailly M,Wyckoff JB,Segall JE,

Condeelis JS.EGF stimulates an increase in actin

nucleation and filament number at the leading edge

of the lamellipod in mammary adenocarcinoma cells.

J Cell Sci1998;111:199–211.

26.Mochly-Rosen D.Localization of protein kinases by

anchoring proteins:a theme in signal transduction.

Science1995;268:247–51.

27.Toker A.Signaling through protein kinase C.Front

Biosci1998;3:D1134–47.

28.Rabinovitz I,Toker A,Mercurio AM.Protein kinase

C-dependent mobilization of the a6h4integrin from

hemidesmosomes and its association with actin-rich

cell protrusions drive the chemotactic migration of

carcinoma cells.J Cell Biol1999;146:1147–60.

29.Parsons M,Keppler MD,Kline A,et al.Site-directed

perturbation of protein kinase C-integrin interaction

blocks carcinoma cell chemotaxis.Mol Cell Biol2002;

22:5897–911.

30.Gauthier ML,Torretto C,Ly J,Francescutti V,O’Day

DH.Protein kinase C a negatively regulates cell

spreading and motility in MDA-MB-231human breast

cancer cells downstream of epidermal growth factor

receptor.Biochem Biophys Res Commun2003;307:

839–46.

31.Ivaska J,Whelan RD,Watson R,Parker PJ.PKC

epsilon controls the traffic of h1integrins in motile

cells.EMBO J2002;21:3608–19.

32.Iwabu A,Smith K,Allen FD,Lauffenburger DA,

Wells A.Epidermal growth factor induces fibroblast

contractility and motility via a protein kinase C y-

dependent pathway.J Biol Chem2004;279:14551–60.

33.Hirai T,Chida K.Protein kinase C f(PKC f):activation

mechanisms and cellular functions.J Biochem(Tokyo)

2003;133:1–7.

34.Joberty G,Petersen C,Gao L,Macara Ig.The cell-

polarity protein Par6links Par3and atypical protein

kinase C to Cdc42.Nat Cell Biol2000;2:531–9.

35.Lin D,Edwards AS,Fawcett JP,Mbamalu G,Scott JD,

Pawson T.A mammalian PAR-3-PAR-6complex impli-

cated in Cdc42/Rac1and aPKC signalling and cell

polarity.Nat Cell Biol2000;2:540–7.

36.Etienne-Manneville S,Hall A.Integrin-mediated

activation of Cdc42controls cell polarity in migrating

astrocytes through PKC f.Cell2001;106:489–98.

37.Etienne-Manneville S,Hall A.Cdc42regulates GSK-

3h and adenomatous polyposis coli to control cell

polarity.Nature2003;421:753–6.

https://www.sodocs.net/doc/a89928815.html,udanna C,Mochly-Rosen D,Liron T,Constantin

G,Butcher EC.Evidence of f protein kinase C

involvement in polymorphonuclear neutrophil integ-

rin-dependent adhesion and chemotaxis.J Biol Chem

1998;273:30306–15.

39.Giagulli C,Scarpini E,Ottoboni L,et al.RhoA and f

PKC control distinct modalities of LFA-1activation by

chemokines:critical role of LFA-1affinity triggering in

lymphocyte in vivo homing.Immunity2004;20:

25–35.

40.Zhang N,Hodge D,Rogers TJ,Oppenheim JJ.Ca2+-

independent protein kinase Cs mediate heterologous

desensitization of leukocyte chemokine receptors by

opioid receptors.J Biol Chem2003;278:12729–36.

41.Chung CY,Potikyan G,Firtel RA.Control of cell

polarity and chemotaxis by Akt/PKB and PI3kinase

through the regulation of PAKa.Mol Cell2001;7:

937–47.

42.Kozikowski AP,Sun H,Brognard J,Dennis PA.Novel

PI analogues selectively block activation of the pro-

survival serine/threonine kinase Akt.J Am Chem Soc

2003;125:1144–5.

43.Parent CA,Blacklock BJ,Froehlich WM,Murphy DB,

Devreotes PN.G protein signaling events are activated

at the leading edge of chemotactic cells.Cell1998;95:

81–91.

44.Leitges M,Sanz L,Martin,P.et al.Targeted

disruption of the f PKC gene results in the impairment

of the NF-n B pathway.Mol Cell2004;8:771–80.

PKC f in EGF-Induced Chemotaxis

相关主题